Comparison and analysis of lncRNA-mediated ceRNA regulation in different molecular subtypes of glioblastoma.

摘要: Glioblastoma multiforme (GBM) is the most malignant brain tumor with a poor prognosis. A molecular level classification of GBM can provide insight into accurate patient-specific treatment. Competitive endogenous RNAs (ceRNAs), such as long non-coding (lncRNAs), play an essential role in development tumors and are associated survival. However, pattern lncRNA-mediated ceRNA (LMce) crosstalk different subtypes still unclear. In this study, we present computational cascade to construct LMce networks investigate lncRNA–mRNA regulations among them. Our results showed that although lncRNAs mRNAs subtype were same, regulation relationships these subtypes. 42.5%, 50.9%, 43.5% 65.0% regulatory pairs classic (CL)-, mesenchymal (MES)-, proneural (PN)- neural (NE)-specific. addition, our study identified 61, 132, 24 16 modules which synergically competed each other for miRNAs CL-, MES-, PN- NE-specific. CL- MES-specific mainly involved biological functions cell proliferation, apoptosis migration, while NE-specific related DNA damage cycle dysregulation. Survival analysis demonstrated some could be potential prognostic markers patients CL MES This uncovered interaction patterns subtypes, subtype-specific distinct functions, revealed These might contribute discovery biomarkers more therapeutic process.